Simultaneous PI for several treatment groups in neg_bin_pi and
quasi_pois_pi (argument “adjust”)
Small change in the test suite for plot function (see issue on
github)
New citation for neg_bin_pi() and quasi_bin_pi()
predint 2.1.1
Small bugfix in the test-suite
predint 2.1.0
New real life data
Mortality of mice
revertant bacteria colonies
new functions
rnbinom
nb_pi
neg_bin_pi
plot works now with uncalibrated PI
predint 2.0.0
the package is now based on s3 objects
the calibration algorithm works now independently for both limits
(default for discrete intervals)
rqpois() can be used with offsets and returns a data.frame (instead
of a vector as before)
quasi_pois_pi() works now with offsets and based on the modified
calibration algorithm
new functions
summary.predint
plot.predint
as.data.frame.predint
print.predint
predint 1.1.1
Argument for the tolerance in the bisection
Small changes in the readme
Bugfix in beta_bin_pi
Changed lambda_min and lambda_max to delta_min and delta_max in
order to reflect the notation of the manuscript submitted to JSS
Bugfix in rbbinom: rho must be bigger than 0 but smaller than 1
Function for the sampling of bs-data based on a random effects model
fit with lme4::lmer() as used in lmer_pi_futmat()
predint 1.1.0
lmer_pi() is now deprecated, but three new functions are introduced
for the computation of PI based on random effects models fit by
lme4::lmer(). The new functions are lmer_pi_unstruc(), lmer_pi_futvec()
and lmer_pi_futmat().